Version classiqueVersion mobile

Pigments et colorants de l’Antiquité et du Moyen Âge

 | 
Institut de recherche et d'histoire des textes
, 
Centre de recherche sur les collections
, 
Équipe Étude des pigments, histoire et archéologie

Session C. Matériaux et traditions techniques : des données pour l'histoire de l'art, l'histoire des techniques, l'histoire économique et sociale

Natural dyestuffs; history of technology and scientific research

Judith H. Hofenk De Graaff et Wilma G.Th. Roelofs

Texte intégral

  • 1 E. PLOSS, Ein Buch von alten Farben, Heidelberg und Berlin, 1962; F. BRUNELLO, The Art of Dyeing in (...)

1As in many other fields of material research into objects of art and science, the research into the identification of natural dyestuffs and pigments should be carried out in collaboration between historians, art historians and scientists. This contribution aims to emphasize the necessity and possibilities of such a cooperation. In one of the Apocrypha about the life of Christ, Jesus is represented as a dyer’s apprentice. In the absence of his master he put some pieces of cloth in the same dye bath, and at the end of the dyeing process each piece has acquired a different colour. A miracle is said to have taken place1.

2The second story is supported by documentary evidence.

3In 1467 Pope Paul II proclaimed that in future the robes of the cardinals were to be dyed with kermes instead of imperial purple. The fall of Constantinople was not the only reason for the Pope’s proclamation: in 1461 a man called John de Castro found in the neighbourhood of Civitavecchia near Tolfo enormous deposits of alum stone. He went to the Pope and tried to convince him of the importance of his findings. But he had to demonstrate several experiments before the Pope was convinced.

  • 2 W. BORN, Der Scharlach, Ciba Rundschau, 7, 1936, 218-243.

4In 1466 the Medici family leased the exploitation of the mines. The Pope gained great financial benefits from this transaction and could hereby pay for the war against the Turks, the former suppliers of large quantities of alum2.

5How can these two stories contribute to the history of technology and, what is their function in scientific research into natural dyestuffs?

6The first story is quite easy to explain for a textile chemist. Most natural dyestuffs are mordant dyestuffs. This means that the colour is formed by a complex of a metal salt with organic material obtained from plant or animal.

  • 3 A.G. PERKIN and A.E. EVEREST, The natural organic colouring matters, London 1918.

7The shade of the coulour depends on the metal used. Madder with iron gives a violet hue. With alum, a brick-red colour is obained and with tin the result is a bright red3.

  • 4 Analysis carried out by the authors, on request of the Hebrew University of Jerusalem

8An historian knowing these facts can better interpret the state of the art of the technique of textile dyeing at the beginning of our era. When these suppositions are supported by investigation of textiles dated 73 AD and found at Masada in Israel, where madder is identified together with aluminium on red samples and madder with iron on violet samples, the picture is complete4.

9Although the second story seems very clear, it may still require more historical explanation. The reason for the Pope’s proclamation could be the financial profit from the alum. This had to be used as a mordant together with kermes to obtain the cardinals’ purple. Imperial as « real » purple (which is obtained from a species of snail) is a vat-dyestuff and does not need a mordant in the dyeing process. However the question can be raised whether imperial purple - a very expensive dyestuff - was used exclusively for the cardinal’s robes.

  • 5 Analysis carried out by the authors on request of the Abegg foundation, Riggisberg, Switzerland. Re (...)

10Analytical research into a large number of ecclesiastical textiles from the cathedral at Sens, from various cloisters in Switzerland, and from treasuries of Cologne, Aachen and Maastricht, has shown that only in a few cases was imperial purple present. Most of the purples in these textiles are dyed either with a mixture of woad and madder, or with kermes. In this case both madder and kermes are mordanted with alum5.

11Textiles of Sicilian origin are often dyed with madder on an iron mordant. The question can be raised whether the Pope’s proclamation merely authorize an already existing practice and at the same time profit from it through the alum monopoly.

12The lesson from this story is that it is not allowed for art historians or historians to classify ecclesiastical purples before 1467 as imperial purples without a scientific identification of the existing dyestuff. The impression that imperial purple is exclusive and reserved only for the very rich can be supported by the following paragraph.

  • 6 K. REINKING, Die Färberei der Wolle im Altertum, 1932 (unpublished manuscript). R. PFISTER, Teintur (...)

13A well known collection of recipes for dyeing are found in the Papyrus Leiden X and in the Papyrus Graecus Holmiensis. These papyri are quite similar. They are both written in a Greek-Egyptian dialect and can be dated as 3rd or 4th century AD. They consist recipes for dyeing gold, silver, stones and purple on textiles6.

  • 7 R. PFISTER, Matériaux pour servir au classement des textiles égyptiens postérieurs à la conquête ar (...)

14The Papyrus Graecus Holmiensis is probably a luxury edition for a priest and not meant for practical work, but the Papyrus Leiden X contains a number of practical recipes for dyeing purple. None of these recipes, however, describes the process of dyeing imperial purple, with Murex snails. They are all recipes for imitations, e.g. with Archil, a lichen which gives a brilliant violet colour, but with a very poor light-fastness, and further recipes for mixtures of woad and madder on an alum mordant. Scientific research has proved the existence of the latter. One of the first chemists who carried out dyestuff analysis, the Frenchman Pfister, investigated a large number of early Coptic textiles from Egypt. He never found imperial purple, - but in almost all cases woad with madder in the purple shades7. These results are proved by more recent analysis using thin-layer chromatography and infra-red spectroscopy.

  • 8 Analysis carried out by the authors on request of the Abegg foundation, Riggisberg, Switzerland.

15Two investigations into textiles belonging to very important personalities prove, however, that imperial purple was used for these people exclusively. In both textiles imperial purple was positively identified. One came from a tomb from the period of the Emperors in Rome. The other came from the tomb in Greece which is attributed to Phillip II of Macedonia. These two practical examples of analytical identification of dyestuffs again support the theory of the use of imperial purple for dyeing and the documentary sources about this topic8.

  • 9 J.H. HOFENK DE GRAAFF, Natural Dyestuffs, Origin, Chemical Constitution, Identification, ICOM, Comm (...)

16In the foregoing, scientific research of natural dyestuffs is mentioned. The question can be raised how does such research come about, what is the starting point and how is it organized? In the Central Laboratory is started about 25 years ago and was caused by historical curiosity in combination with knowledge of textile chemistry as it stood. At first we followed in Pfister’s footsteps, later discovering modern methods of chemical analysis. The presence of a colleague specialized in chromatographic techniques makes the picture complete9.

17In our case a symbeosis was created between chemical, textile-technical and historical knowledge.

18In the beginning the analytical methods of the identification was perfected and the fact that nowadays more possibilities are available can be learned from lectures during this conference.

19However, when the moment comes that the analytical method can be used as a tool for further research, the question arises of how we should continue.

20There are several possibilities, one is to use the analytical method to investigate all the chemical components of a particular dyestuff or pigment, or to use different methods to identify the chemical structure of one or more component. This research can give much information on structure and origin of an unknown dystuff.

21Another possibility is to be satisfied with the identification of the main components (without neglecting the unknown components) and identify a larger number of textiles to answer a historical question.

22In the Central Laboratory we have chosen the second way of continuing the research.

23The first research investigated the substitution of kermes by cochineal in the 16th century. These two dyestuffs are both obtained from a scale-insect. Kermes originates from Southern Europe and North-Africa. Cochineal comes from South America. Kermes contains mainly kermesic acid, cochineal mainly carminic acid. The two can be separated by chromatographic methods. The method of dyeing and the colour obtained is almost identical.

24In various historical sources it is mentioned that cochineal replaced kermes very soon after the discovery of America by Columbus. The dominicans, whose province in the sixteenth century broadly coincided with south eastern Mexico, are known to have stimulated production from 1540. Grana (cochineal) was incorporated into the royal tribute systems by at least 1530.

25The total amount of cochineal marketed through official and private channels from the mid 1530’s was probably substantially in excess of that circulating the 1520.

26To support this by scientific research, hundreds of red textile samples from the period 1450-1600 were collected and the dyestuff identified.

  • 10 J.H. HOFENK DE GRAAFF, W.G. Th. ROELOFS, On the occurrence of red dyestuffs in textile materials fr (...)

27Besides a number of interesting facts it became clear from this research that cochineal was indeed mainly present after 1550 and kermes was the main dyestuff before that date. So the general conclusion was drawn that, after 1550, cochineal replaced kermes almost completely10.

28Revisiting the results after fifteen years we would make a few remarks and restrictions.

29First of all the selection of the textiles under investigations. This selection is very arbitrary. In museums, where the samples came from, mainly very important or rich textiles are kept. This selection in general does not show a representative picture of the daily life in the past.

30Secondly the dating of an object, except for tapestries with a date on it, is often unsure and depend on stylistic elements. Thirdly, the type of textile fibre, e.g. wool or silk, makes quite a difference in the use of a particular dyestuff. It is not possible to give a general statement for both fibres.

31In studying historical sources, such as guild regulations, orders and prohibitions by local authorities, it is clear that many of these are often caused by abuses in daily practice and it is very dangerous to take these guild regulations as the only guide for the history of technology.

32Regional differences must also be taken in account and speak against a general interpretation of the results.

33An example is the use of kermes in textile centres in the Northern Netherlands.

  • 11 J.H. HOFENK DE GRAAFF, Veranderend kleurstofgebruik in de Leidse textielververij in de zestiende en (...)

34From written and printed sources it became clear that kermes was not used at all before the end of the 16th century and that cochineal was used only after 158011.

35The foregoing in a plea to revise the analytical results, together with the most recent historical knowledge. Since the introduction of computers and data-systems new approaches have been made available.

36In the past twenty years several thousands of textile samples have been investigated and the dyestuffs identified. As mentioned before, four hundred red textile samples from the period 1450-1600 have been analysed. However more than six hundred yellow samples and about 1500 samples from tapestries have also been analysed, together with an unknown number of various other textiles.

37The results of all these analyses are recorded, and were samples remain, these are carefully stored.

  • 12 Reflex,Borland/Analytica Inc, 1985.

38With the results of the research on red dyestuffs from 1450-1600, an experiment was started using a data-system called Reflex. This system is chosen because of its numerous search possibilities12. It is easy to operate and inexpensive. It is compatible with Dbase III plus, a system which is used internationally.

39Before setting up a database, you should decide what you want to ask such a system. In our case not only the dating is important, but also the combination of dating and origin. The chemical constitution of the identified dyestuff also the common name of the dyestuff should be introduced. The owner, type of textile, fibre material and all this is combination with each other.

40An experimental record is drafted which contains all these elements. It contains as much as possible of the data found through the analytical identification and the historical information (fig. 1). The analytical data are defined by our method of analysis thin-layer chromatography. However this can be extended to results of other analytical methods such as metal analysis of the mordant, or results of high-performance liquid chromatography, or Fourier Transform Infrared Spectroscopy. Because it is an experiment we hope that colleagues will comment our records and system.

Fig. 1. Design of a record structure for analytical results in the Reflex Database system.

41As an experiment we have introduced the data from the already mentioned red dyestuffs. During the introduction it became clear that much of our data is incomplete and this is a warning for the future to write down as a much information as possible (fig. 2).

Fig. 2. A filled out record in the datasystem Reflex with the results of an investigated object.

42The introduced data can be arranged through the search programme according to various points of view.

43A possible request might be: “give all textiles from the 3rd quarter of the sixteenth century”. A list is created of all the textiles which fulfil this condition. Or: “give all Spanish textiles, from the first quarter of the sixteenth century, dyed with madder”. A new list is created (fig. 3).

  • 13 Analysis carried out by the authors on request of the Centro Restauro Manufatti Tessili, Milano.

44Last year we have asked the system various questions and combinations of questions and the results were remarkable. An example is that in this group of textiles red Italian silks from the 15th and 16th century were dyed mainly with so-called Polish cochineal, a scale-insect, which contains a mixture of carminic acid and kermesic acid13.

Fig. 3. Example of a list of the results after using the search conditions of the Reflex System.

  • 14 Analysis carried out by the authors on request of the Royal Library, The Hague.

45This result gave us the idea of using our collection of existing samples as reference material. An opportunity was given by a request for scientific investigation of the dyestuff from a late mediaeval “chemise”. This chemise, a red silk velvet wrap of an illuminated manuscript, probably from the 15th century, could be Italian or Spanish. A question on dating was also raised: was it late fifteenth or at least early sixteenth century, or even much later? The last question is always difficult to answer but worth a try. How to tackle this problem14.

46We asked the data-system the following question:

47Which textiles from our sample collection fulfil the following conditions: red; velvet; silk; Italy; sixteenth century; dyed with Polish cochineal?

48The computer gave the answer: five textiles fulfil these conditions. These five textile samples were used as reference material together with the unknown sample. The analytical results showed the presence of mainly carminic acid with a minimum quantity of kermesic acid and Brazilein. The interpretation in relation with the dating was difficult, but indicates cochineal with Brazil wood. After more detailed information from the owner it was indicated that the manuscript was probably Spanish. Through the search programme it became clear that the presence of cochineal in Spain dates from an earlier period than in Italy.

49Dating of a textile with the help of a dyestuff analysis remains difficult and more research is needed into chemical components of the various scale-insects.

  • 15 J. WOUTERS, High Performance Liquid Chromatography of Anthraquinones. Analysis of Plant and Insect (...)

50We would like to give this problem in the hands of those colleagues who are specialized in methods which can identify the various components of a particular dyestuff in more detail15.

51Historical evidence of the use of various scale-insects is available in various dye-recipe books.

52One of the first recipe books on textile dyeing is the Plictho de l’arte de Tentori by Rosetti, and printed in Venice, in 1548.

  • 16 G. ROSETTI, The Plichto, Translation of the first edition of 1548 by S.M. EDELSTEIN and H.C. BORGHE (...)

53In one of his recipes for dyeing silk red, he mentiones seven various scale-insects varying in quality, including crimson of the March, of the Levant, of the West, crimson Slavic or Ragusian and various qualities of these16.

  • 17 N.W. POSTHUMUS, W.L.J. de NIE, Een handschrift over de textielververij uit de eerste helft der zeve (...)

54But Rosetti does not stand alone. In an early 17th century Dutch manuscript on textile dyeing, the following types of kermes or cochineal are mentioned17.

55The best is Turkish of Syrian, followed by Spanish and French, but the poorest is that from Germany. Amusing is the remark that in Italy and France these various qualities are mixed together. Only when the textile is intended for high-ranking persons such as emperors, kings or princes, and for the cardinals who always wear scarlet robes, do they discriminate between the various qualities. Typically Dutch might be the remark that when the textiles are intended for such persons, they should pay for it.

56As you see, here is a need for further research and are there possibilities for cooperation between historians, scientists and perhaps even botanists.

57A few more examples can illustrate our approach.

58At the beginning of the seventeenth century, Indian printed cottons (chintzes) were imported into Europe. Soon after the introduction it became an important product for the East India Companies.

  • 18 J.H. HOFENK DE GRAAFF, Sitsen en hun Europese navolgingen, 1600-1800. De techniek ontrafeld!, Rijns (...)

59Imitations were made in Europe and European patterns were exported to India to be produced there. The dating and identification of the origin of those textiles sometimes gives problems. Through historical research into written sources from the 17th century, both from India and from Europe, the technique of painting and printing of the chintzes could be reconstructed. Criteria could be formulated to distinguish the Indian from the European chintz. One of the criteria was the presence of Saya wera, a madder-type dyestuff which was used exclusively in India. This dyestuff contains only alizarin whereas the European madder contains both alizarin and purpurin. This phenomenon was checked by the chromatographic analysis of the red parts from various chintzes, and found to be correct18.

60The last case contains a Persian rug.

61The main colour of the rug was red, but the upper part of the rug was brownish red, the lower part bright red. The question coming from the curator was:

62Where these two colours originally intended to be different, or is it a question of fading. After having analysed the two parts it became clear. The upper part consisted of a minor quantity of cochineal and a large quantity of Brazil wood. The lower part contained only cochineal. As it is known that Brazil wood fades under the influence of light to a brownish hue, the problem could be solved.

  • 19 Analysis carried out by the authors on request of the Rijksmuseum in Amsterdam.

63The colour of the rug was consistent at its production, but because two types of dyed wool were used these faded differently19.

64There are many more examples to demonstrate, but we hope that through this contribution we have given an introduction to our approach and hope that it will stimulate other colleagues, to work in this field.

Notes

1 E. PLOSS, Ein Buch von alten Farben, Heidelberg und Berlin, 1962; F. BRUNELLO, The Art of Dyeing in the History of Mankind, Vicenza, 1973.

2 W. BORN, Der Scharlach, Ciba Rundschau, 7, 1936, 218-243.

3 A.G. PERKIN and A.E. EVEREST, The natural organic colouring matters, London 1918.

4 Analysis carried out by the authors, on request of the Hebrew University of Jerusalem

5 Analysis carried out by the authors on request of the Abegg foundation, Riggisberg, Switzerland. Results are published in; Bulletin de Liaison CIETA, 1976, 43-44; 101-111 and M. FLURY-LEMBERG, Textilkonservierung im Dienste der Forschung, Bern, 1988.

6 K. REINKING, Die Färberei der Wolle im Altertum, 1932 (unpublished manuscript). R. PFISTER, Teinture et alchimie dans l’Orient hellénistique, Semenarium Kondakovianum, VII, 1935, 2-59.

7 R. PFISTER, Matériaux pour servir au classement des textiles égyptiens postérieurs à la conquête arabe, Revue des Arts asiatiques, Tome X, Fascicule 1, 2, Paris, 1936, 1-15, 73-85.

8 Analysis carried out by the authors on request of the Abegg foundation, Riggisberg, Switzerland.

9 J.H. HOFENK DE GRAAFF, Natural Dyestuffs, Origin, Chemical Constitution, Identification, ICOM, Committee for Conservation, Brussels, 1967; second version, Amsterdam 1969.

10 J.H. HOFENK DE GRAAFF, W.G. Th. ROELOFS, On the occurrence of red dyestuffs in textile materials from the period 1450-1600. ICOM Committee for Conservation, Madrid, 1972.

11 J.H. HOFENK DE GRAAFF, Veranderend kleurstofgebruik in de Leidse textielververij in de zestiende en zeventiende eeuw, Rijnsaterwoude, 1984.

12 Reflex,Borland/Analytica Inc, 1985.

13 Analysis carried out by the authors on request of the Centro Restauro Manufatti Tessili, Milano.

14 Analysis carried out by the authors on request of the Royal Library, The Hague.

15 J. WOUTERS, High Performance Liquid Chromatography of Anthraquinones. Analysis of Plant and Insect extracts and Dyed Textiles, Studies in Conservation, 30, 1985, 119-128.

16 G. ROSETTI, The Plichto, Translation of the first edition of 1548 by S.M. EDELSTEIN and H.C. BORGHETTY, London, 1969.

17 N.W. POSTHUMUS, W.L.J. de NIE, Een handschrift over de textielververij uit de eerste helft der zeventiende eeuw. Ned. Econ. Hist. Archief, Jaarboek XX, 1936, 230-234.

18 J.H. HOFENK DE GRAAFF, Sitsen en hun Europese navolgingen, 1600-1800. De techniek ontrafeld!, Rijnsaterwoude, 1986.

19 Analysis carried out by the authors on request of the Rijksmuseum in Amsterdam.

Table des illustrations

Légende Fig. 1. Design of a record structure for analytical results in the Reflex Database system.
URL http://books.openedition.org/editionscnrs/docannexe/image/8173/img-1.jpg
Fichier image/jpeg, 58k
Légende Fig. 2. A filled out record in the datasystem Reflex with the results of an investigated object.
URL http://books.openedition.org/editionscnrs/docannexe/image/8173/img-2.jpg
Fichier image/jpeg, 86k
Légende Fig. 3. Example of a list of the results after using the search conditions of the Reflex System.
URL http://books.openedition.org/editionscnrs/docannexe/image/8173/img-3.jpg
Fichier image/jpeg, 342k

© CNRS Éditions, 2002

Conditions d’utilisation : http://www.openedition.org/6540

Cette publication numérique est issue d’un traitement automatique par reconnaissance optique de caractères.
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search